I'm Getting into Mesh Networks (Meshtastic, MeshCore, and Reticulum)

> the author also misses an important piece of functionality of meshcore: If I lose power, the mesh still works.

Isn't it the case with Meshtastic and Reticulum too? It feels like it should be part of the definition of mesh network.

> no one can stop me or control what is said no matter what

Can't they just triangulate the nodes and hack or unplug them? And put whoever objects into prison?

> but I really feel that these fully decentralized solar powered networks are hugely important as a simple alternative to the corporate behemoth the internet has become.

Is the internet that? People have built corporate behemoths on top of it, in world wide web-land, but the internet itself seems relatively neutral.

The meshcore software, and the common hardware being used, are both comically weak for anything that approaches usage at scale, especially in an emergency situation.

The range is extremely limited, and the throughput gets really bad if your packets have to travel more than a few hops. These two factors alone combine to write this off as nothing more than a toy pretty much from the start.

There are already unlicensed radios with longer range that would be a better starting point if people were trying to position mesh* as a scalable and reliable transport of any kind.

>Second, the author also misses an important piece of functionality of meshcore: If I lose power, the mesh still works.

This point isn't unique to meshcore, and it is not a guarantee. Any solar powered and battery-backed device can function without utility power (in theory). Meshcore nodes are not solar powered by default, and the same solar power concept can be utilized for any other kind of radio transceiver/protocol.

Meshtastic simultaneously has too few nodes for most people to see many reliably, and has a scaling problem where too many nodes where saturate the network.

I ran some Meshtastic nodes for a while, same overall experience.

Rarely saw nearby nodes, never communicated anything more than a basic "HELLO"/"ACK" kind of thing.

It's a neat idea for things like a distributed sensor network on your own property, or other IoT kinds of comms. It's not a practical platform for human to human comms, especially in a disaster scenario.

All ham radio repeater groups here dropped Meshtastic as it was super unreliable. And they know how to build proper antennas and filters.

Meshcore is 100% free. The last issue was the closed sorce Android/iPhone client - but there are FOSS Flutter based-Opensource clients available (https://github.com/zjs81/meshcore-open)

I really like the technology and almost jumped in, but quickly realized the lack of the _network_ effect in a community.

I live in a dense urban area, and yet I see very few nodes nearby, where I image I will be left in starvation. Without a proper mesh, I couldn't send a message to a fried 2 miles away with no los.

Have you tried MeshCore? It seems there's a lot more people on it (at least here in Europe).

I believe it's called ham radio

> but without enough structure (or mass appeal) to take off

I think the most important requirement for a mesh technology to take off is a purpose that is practically relevant right now. Let's assume the mesh network exists - now what? Now you can send messages to other nerds but ... what do you even want to send as a message? That's why ham radio basically bottomed out at contests, Morse code challenges and exchanging specs ... there is nothing to say. Maybe the biggest problem of mesh networks isn't technology but society. If there is a purpose that at least serves nerds making up 0.1% of the population then that would be amazing and mass appeal would actually be more cause for trouble than desirable.

LoRA can work but there needs to be a backbone, either terrestrially via towers with repeaters or LEO constellations. Everything else is doomed to fail.

I had a friend who worked in telco as well, and he told me the reason why IP-based networking solutions will never completely replace earlier stuff like GSM is that the QoS algorithms used work extremely well in practice (which mesh networks prove), their behavior is completely nondeterministic, and tends to have really bad failure cases in overload, or certain nodes in the network failing.

Which means in SHTF situations (such as one when most people imagine this would be useful), these things tend to fail the most.

That weakness is a strength.

Everyone you meet on a mesh is a real breathing nerd, who due to proximity has a lot in common with you. They are not trying to influence you or sell you anything

How many places like that are left?

The inherent limitations of free spectrum mesh technologies will never lend itself to a replacement for the Internet so will always largely exist in niches. Niches like personal nets, local nerd networks, or emergency response (tho actual first responders are not the most eager to try this stuff based on my experiences in the community.) All of this can be a feature or a bug depending on whom you ask.

This is by design. It’s like BBS again

It's pretty hard to imagine anyone "getting serious" with these tiny radios designed for reading gas meters and weather stations and using that to build some kind of off-grid alternate Internet.

I'd almost have more faith on dragging out all the old acoustic coupler modems and building a city-wide string-and-tin-can telephone network.

Of course they are not useless. My hiking/camping friends put together a fun orienteering game which used Meshtastic to do live GPS tracking. Worked great for that. But a country spanning meshternet it is not.

I don't think it needs to be though. There are plenty of things to explore using these things at the local 1-10 km scale.

To be fair it is already a miracle that there is enough Metastatic in my area that I can (sometimes/rarely) send a message between my home and work!

*disclaimer that I am coming off a recent disillusionment with Meshtastic. I thought it would be fun to make a single raspberry pi image with all the dev tools on it to do some off-grid dev/maintenance work if you were to treat this seriously and pretend the main internet is down. That moment came when I tried to compile something and the Pi ran out of memory. Really? I need more than 1GB RAM to compile something used to send short text messages?! That's nuts!

On the mesh in Toronto with meshcore we have regular communication that reach all the way to Buffalo. We are past the « toy » stage, it’s truly impressive.

I'm right now anchored at an atoll in the Tuamotus, French Polynesia. 3/10 boats anchored here have Meshtastic.

I think meshes do work extremely well in practice, and are quite resilient with regards to errors, and load balancing, and they get better as you add more nodes.

I think it's perfectly feasible for a small neighborhood of regular people to have internet shared over a wireless mesh network, yielding experience comparable to standard approaches.

In Montréal we've rebooted Réseau Libre which used to be a Wi-Fi mesh experiment 15+ years ago. It's a fun experiment, but in a way feels like a step backward for me. Meshtastic and Meshcore are just that, messaging, but that makes it the standardized killer app. On the other side you have reticulum which allows decoupling from the LoRa low bandwidth only radios, seems to do a lot of neat stuff, but if we're reinventing a whole network layer, we're going to have to reinvent services, discovery, etc. and I fear we're wasting time when in the end what wins is controlling the backbone bandwidth, but with the added difficulty of a p2p mesh.

I'm starting to feel this is a fun activity, but realistically copium for a world that is very sadly centralizing everything.

The issue is that these mesh protocols quickly break down under any real loads.

Setting up a few mesh nodes, running some tests, and thinking you have a kit that is usable in an emergency is like so many other "disaster recovery" drills we've been through that assume ideal conditions. The excellent daily tape backups that you realize too late you can't utilize in a bare metal recovery situation because nobody kept an OS install media handy, or they forgot to keep the installer and license keys for the backup software in the datacenter.

The challenge with these mesh systems is that few, if any, areas have even gotten to a point that they could run a realistic simulation of relying on this system for communications.

What do you need 200 miles of range for?

Not having great luck with my solar node. Nothing ever seems to get out, even with a pretty nice antenna...

Did you set up solar nodes in random spots or only areas you own/control?

I've wondered if it'd be legal to just chuck a couple into some random trees.

And when the temperature drops below zero?

I wonder what you are suggesting cell phones should be able to do here?

These ideas yes, but these networks already have a concept of message oriented semantics and so there's not much of a need to rebuild most of those protocols. A lot of what Finger, Gopher, et al does is define the application layer semantics to transmit documents over stream oriented protocols.

A lot of the "early internet" protocols would likely pair quite well, especially ones based on UDP.

However, Reticulum has its own active "small web" implementation with NomadNet and the Micron markup.

Isn't that exactly the idea of Reticulum?

http://reticulum.network/manual/whatis.html

Do you know what the MSS is on a Halow network? Curious if it makes sense to run usual TCP and UDP based applications on them or whether we would need to switch to things like CoAP.

I think there are people playing with Halow with Reticulum. That's one advantage of having a multi-transport system. There is also a Bluetooth transport now: https://salemdata.net/johnpress/?p=720

I mean morse code on GMRS is actually an amazingly strong physics solution. Take the benefits of VHF propagation and combine it with high power limits and a coding scheme that is on par with FT8 for noisy channel resilience. No way a potato powered microwave is going to compete.

915 MHz mesh isn't a fair comparison. APRS is, but that requires licensing and unencrypted communication, so it gets less traffic. Quite good and fun though. I get point to point pings dozens of miles away daily.

it works incredibly well at Burning Man, so you are completely right!

Uhhhh ... No. You must have read that it uses the same frequency 900mhz, but did you actually try using it? When I first got on meshcore here NJ i immediately connected with my closest neighbor repeater 20 miles away which then in turn connected me to the local NJ/PA mesh which spans almost 200 miles wide. I don't recall any cordless telephone ever doing that...

WCMesh in California covers a few hundred miles of southern California on Meshcore.... That isn't flat. It really just depends on buy-in in the local region.

I have a very different experience - here in Europe with 868 MHz MeshCore you get good singnal from a repeater through one or two city blocks with non ideal antenna placement.

With reasonable line of sight tens of kilometers & much more is doable. There are some repeaters on mountains that connect bigger regional meshes with packets going >100 km regularly.

https://mapa.meshcore.cz/

For emergency comms(remember, emergency services provided by the state don't monitor the ham bands, so can't call them), a $20 ham radio would work or just carry FRS radios.

FCC does allow use of a ham radio in a real emergency without a license.

None of MeshCore, Meshtastic or Reticulum will scale well, especially not on top of a heavily constrained radio technology like LoRa. Flooding is inefficient for obvious reasons, AODV-esque routing (which MeshCore tries to do for DMs and Reticulum tries to do in general) is prone to almost-immediate path failure on unstable underlying transports, and the hidden node problem always bites on haphazard/unplanned mesh radio deployments where people show up with nodes in random locations on the same frequency.

The cracks are already extremely visible in MeshCore in the UK, where overheads from adverts and dropped packets from collisions mean it is already horrendously unreliable and most of the chatter in the Public channel is people sending test messages and being unsure whether anything they sent was ever heard by anyone.

Most other routing protocols (BATMAN included) are also not that well suited to situations where the underlying transport ends up asymmetric, e.g. one node can't hear others but it can be heard, and that's an extremely common occurrence/failure mode in wireless meshes like this. It's a difficult problem to solve with coordination between nodes, let alone without.

You can build one with an SX1302/SX1250 and an STM MC for cheap.

Official manual including install instructions: https://markqvist.github.io/Reticulum/manual/

Reticulum would probably be the best one for it. As far as I can tell there aren't any interfaces for it already, but if you can get a command that transfers data using stdin and sdtout, Reticulum has things like the PipeInterface https://reticulum.network/manual/interfaces.html

I have some really bad news for you, mate.

No, you're not alone. MeshCore is very neat tech, but I love that you can show up to, say, a music fest with a few Meshtastic radios and voila, instant mesh. To me, it feels more in the spirit of the thing. That's purely subjective, but that's how I see it.

You totally can use Meshcore without repeater, and companions can be used like routers. This functionality can be enabled any time using one checkbox in app. It switches the radios to slightly different frequency and enables repeat mode in personal nodes, this gives you one small network. No repeater needed. And in this mode you don't clutter "main" mesh with your local mesh.

The main difference between Meshcore and Meshtastic is how telemetry is handled. In Meshcore to get telemetry the other party needs to request it, whereas in Meshtastic telemetry is sent in flood mode in configured intervals. That's why Meshtastic is better suited for (A)TAK [1]. But because Meshtastic sends telemetry anyway there is less and less airtime for chat messages, and it gets to the point where you can't talk to people. For small groups this is fine, for bigger groups/meshes this is no bueno.

[1] https://tak.gov/ - (A) stands for Android app.